I have an issue where on the second time of displaying this interface tapping on the textField will cause the app to crash. How should I proceed in when debugging this crash, is there anything in the log to help me pin point the issue?
After I removed dealloc method crashes disappeared. Therefore I decided to find new way to implement autosuggestion feature.
The interesting thing is that I didn't import header file in code anywhere, but crashes occurred.
Related
Further to my earlier question
Migrating iOS Hybrid App from UIWebView to WKWebview
I have made good progress and only have a few loose ends to tidy up. The overall performance improvement over UIWebView is outstanding.
In UIWebView it was possible to set the focus on a text field programatically using
webView.keyboardDisplayRequiresUserAction = NO ;
However this is not available in WKWebView and ever since 2016 programmers have been developing an updating work arounds (swizzles) to overcome this.
I have tried implementing the latest of these I could find on Stack Overflow, which I found at https://stackoverflow.com/a/55344531/5948260
However as I am very inexperienced at coding in XCode please could someone tell me exactly how I add the above solution to my project. I have tried in vain to find examples of how exactly to do this i
on GitHub or elsewhere but to no avail.
So far I have tried copying the code provided in the above answer into a .m file and adding it to my project, but I got 20 or so errors, I then added an import statement for Foundation and most of these went away but XCode complained that it did not know what class WebViewInjection is. Also must there be a corresponding header file?. Must there be a corresponding interface statement? How does the method defined in the answer get invoked?
As I could not answer any of these, I also tried adding the method into my ViewController class. Whilst this was accepted with no errors, it had no effect on my app, ie. the keyboard did not come up automatically.
I would also want the webview to resize to appear above the keyboard and not to scroll to where the text field is as this seems a very messy solution to me.
All help very gratefully received.
I have now managed to get the keyboard displayed by adding
[self allowDisplayingKeyboardWithoutUserAction];
to viewDidLoad
I kept the allowDisplayingKeyboardWithoutUserAction method declaration inside my ViewController but changed the + infront of the method declaration to a - and this seems to have worked.
I still do not know how to resize the webview when the keyboard is displayed.
I have a Notes-like app: UITableViewController showing up individual notes by pushing them onto navigation stack.
The problem arises when I have UITextView with the FirstResponder status (keyboard is shown) and I touch Back button. The current view controller is dismissed with the animation as expected, BUT the navigation bar is broken now! If I press any of the bar buttons, it will cause EXC_BAD_ACCESS.
I would say that it was not transitioned properly. The table VC is broken somehow as well, as it may appear empty on further manipulations... Very strange behaviour!
By the way, it did not cause any problems with iOS5 and iOS6, but there I used a custom chevron Back button.
I've checked the standard Notes app and it works like a charm.
What is the trick?
Thanks a lot for your advice!
EXC_BAD_ACCESS means you are trying to access an object that has been deallocated. Best thing you can do to trace this is enabling NSZombie, it will tell you what released object is being sent a message (aka EXC_BAD_ACCESS).
You can get how to enable it from here.
I got it and will try to explain to help somebody else to save their day...
EXC_BAD_ACCESS was raised because UITableViewController was not properly transitioned to during Back pop-animation (its viewWillAppear: and viewDidAppear: method were not triggered at all).
In its turn, the animation was not properly performed, as popViewControllerAnimated: was called twice or even more times: 1) as part of the system Back-button callback; 2) inside textViewDidEndEditing: in case no text was entered.
The solution is to check whether the back button has been pressed before calling popViewControllerAnimated:. The trick is to check if the detail-view controller is still in the navigation stack.
Here is the helper method:
-(void) returnToTheListOfRecords {
self.textView.delegate = nil; // this is to avoid the second call of `textViewDidEndEditing:`
if ([self.navigationController.viewControllers indexOfObject:self.delegate]==NSNotFound) {
// Back button has been pressed.
} else {
[self.navigationController popViewControllerAnimated:YES];
}
}
The problem happens on iOS7 only due to its brand-new animation.
I have the following scenario:
UIViewControllerA pushes UIViewControllerB on to the navigation stack via a storyboard segue;
UIViewControllerB is now currently onscreen and I simulate a memory warning via the iOS Simulator.
UIViewControllerA.DidReceiveMemoryWarning() gets invoked as well as UIViewControllerA.ViewDidUnload().
In method UIViewControllerA.ViewDidUnload() I call ReleaseDesignerOutlets() as recommended. UIViewControllerA designer outlets are now deallocated and null.
I hit the "back" button on the navigation bar which dismisses UIViewControllerB and causes UIViewControllerA.ViewDidLoad() to be invoked.
At this point I would expect the designer outlets to be initialised but they are still null and my application crashes due to null pointer exception.
Have I fundamentally missed something? I am using the Monotouch 5.4.0.
EDIT
I found this post which describes the solution to problem to be the same as what I discovered, that is, don't release the designer outlets in ViewDidUnload().
Update: It's finally crash free! (or at least it seems to be)
For future reference, it looks like the big lock-up was due to my disposing of my UIWebView outlet in the ViewDidUnload event. The UIWebView is defined in the xib, and not added as a subview in code.
When memory was running low, the view unloaded and removed the UIWebView outlet. Subsequently, the ViewDidLoad event would fire again, as the app recovered (!?), but it would lock-up. I expected the UIWebView to be reinstated in when the view reloaded (i.e. the xib reloaded).
Is this not the case? I thought it was good practise to dispose of outlets in the ViewDidUnload event?
As soon as I commented out the "webView.Dispose(); webView = null;" line of code, it's been working as expected.
What a day!
EDIT
Further debugging reveals that SOME designer outlets are getting reallocated such as UIImageView and UIScrollView, others such as UITextField and UIBarButtonItem are not.
I'm currently working on a program using MFC. The current third party function starts a thread after an action has been completed using MFC (ie. Checking a checkbox, which starts a MFC thread I believe).
The problem occurs when I check the checkbox, at which point the entire program hangs. I read a few interesting discussions on CProgramming and msdn, it seems that the problem occurs because the new third party thread is calling WaitToSomething() when MFC is updating a control.
Something interesting to note:
When I debug the program, the program hangs (aka. repeatedly calls WaitToRead() ) after I check the checkbox and a new thread is trying to start
When I run the program without debugger, the program is fine UNTIL I switch to another window (ie. Internet browser, Notepad, etc)
My hypothesis:
check to make sure that MFC has finished updating the control before starting a new thread
If anyone has any suggestions or solutions, please leave a comment. Thanks.
Edit:
MFC is not thread-safe at object level (only at class level), so problem occurs when two threads work on the same CButton object.
Q: How do I make it thread safe?
A colleague helped me figure out what the problem was.
The reason why it was hanging is because that the control containing the checkbox is a child dialog, and when it finished updating the message never got passed up to its parent (so when 3rd party thread calls WaitFor(), the MFC thread never completes because a parent dialog thinks its child is still updating the controls).
Fix:
Under 'Properties' in the child dialog's control, set the 'Control' flag to true (and if it has children, set the 'Control Parent' flag to true as well).
Hope this helps.
I've written a Blackberry appliation with the Blackberry JDE, running on a 9000 simulator. I tested it a week or so ago and loaded it on a Blackberry 9000 phone, and everything worked just fine. Sometime between then and now, though, something went wrong.
My code does the whole moving arrow "loading things from the internet" or whatever thing, but no screens pop up. My original screen, which is just a MainScreen with a RichTextField doesn't load at all. This screen, at least, has most likely not changed in the passing week, so if something broke, it would be in one of the later screens/lines of code that it shouldn't even be getting to yet!
Is it possible that my .jad or .cod file are corrupted somehow? I noticed that when I first put code on my machine, I just stuck in the .cod file that Eclipse provided me. Then, last week, the .cod file it gave me didn't work, because it was ACTUALLY a zip file with a two .cod files inside of it. Using the .cod file with the same name as the .cod file they were in succesfully loaded my app. I did the same this time, and I don't get invalid cod file errors or anything, but the app is still as broken.
Is there some direction I should be looking? Is the issue likely to be in my code, the cod file, the phone, or somewhere completely else?
-Jenny
Edit: I've narrowed it down to the problem only occuring if I attempt to load a particular screen. My problem is that this screen is nearly identical to another screen that IS working just fine on the actual device. Both screens are generated from the same method (which makes a webservice call and gets XML back and parses it to populate the fields of the screen). The only difference is that the screen that is breaking is going to a different URL. This URL DOES work (both from a browser and from the simulated device), so I"m at a loss. The application doesn't seem to crash, (it's still running in the background), it just doesn't attempt to display anymore.
Edit:
Okay, I'm seeing some tunneling errors immediately after I load my app, (but before I execute any of my networking code). When i do execute my networking code, it works just fine, unless it happens to be for my "Rental" section. I commented out all calls to that, and made my menu item for Rentals simply make a print statement. The code behaves identically (it freezes, or displays a white screen after selecting the button). All other menu items work (including those that call threads or network methods). And the rentals menu sucessfully executes in the simulator.
private MenuItem _rentals = new MenuItem("My Rentals", 110,
10) {
public void run() {
//if the last thing I did was a rental
//just show the screen
//else, reload rentals
System.out.println("Rentals was selected");
displayError("Rentals was pressed");
// if(rental){
// System.out.println("It's a rental!");
// popScreen(getActiveScreen());
// pushScreen(_offeringsScreen);
// }else{
// System.out.println("Getting Rentals from scratch");
// RentalsThread _rThread = new RentalsThread();
// _rThread.start();
// }
}};
I'm at a complete loss here: The device debugger doesn't seem to even register me selecting the menu item, and not a single line of code executes! It just freezes! I'll try putting back in my RentalsThread call in the start of my program (which was also freezing) just to see if I can tease apart the problem with the Rentals Thread (which makes the Rental Screen), and the problem with the Rentals menu item.
Okay, I think I have this figured out.
1.) My code was still behaving identically even after commenting out everything because I wasn't rebuilding the .COD files (they automatically rebuild if you try to run it in the simulator, but don't when you're generating a .ALX file, for some reason).
2.) The code I had for generating the Rental Screen was adding things to said screen. Apparently this is all well and good on the simulator, but on the real device it's required that you do all graphics manipulation (even for graphics not yet displayed) in an event thread (I used invokeAndWait).
So, now everything seems to be working just fine. There wasn't anything wrong with my networking (nor did I think there was, because my other networking screen works just fine). I still don't know why I get all those weird tunneling network things before I start, but it doesn't seem to affect anything yet.
See also:
BlackBerry UI Threading - The Very Basics
BlackBerry threading model
several suggestions:
if you have some background work with resources like file IO or networking, app just may stuck there... provide error handling and try to debug app from device!
code signing, check latest code update for API which require signing. But since there are no errors this is doubtful.
To debug on device, run Blackberry Device Manager, attach phone to usb, in eclipse select project, Context Menu -> Debug As -> Blackberry Device.
See A50 How to Debug and Optimize
UPDATE I see "Tunnel failed" exception, so it's like network connection problem...
See tunnel failed in blackberry bold. why?
How to Configure Full Internet Access On BlackBerry
UPDATE Support - Application stops responding when opening a connection